A review of basic withdrawal stratagies in retirement.

[Markola]

I think the Kitches article that samclem provided is the best evidence I've seen that simply drawing proportionately from each asset class and -importantly- rebalancing protects the portfolio best. Even simpler would be owning one balanced fund and selling shares of it. No way one needs an expensive FA or fancy formula for that.


Sent from my iPad using Early Retirement Forum

[pb4uski]

+1 except in many cases people will have taxable, tax-deferred and tax-free accounts and a balanced fund is not sufficiently tax efficient.

[samclem]

Quote:

Originally Posted by pb4uski +1 except in many cases people will have taxable, tax-deferred and tax-free accounts and a balanced fund is not sufficiently tax efficient.

That would make an interesting (and very complicated) study--compare the real tax "cost" of owning just a balanced fund (target date, etc) in the typically available taxable, tax-deferred, and tax-free accounts vs optimizing where each holding goes. It would be difficult to make any generalizations because it depends so much on the individual's present and future tax rates and the growth of the portfolio. But I sometimes feel we make too much of it. For example, if a person does tax-loss harvesting in taxable accounts, it feels like a significant tax savings is occurring. But, the cost basis on those shares is reset to a lower amount, so eventually (if the asset is sold), a higher amount will be taxable. If that occurs after the onset of the "tax torpedo" years, then the person could actually be worse off. Likewise, when we move our bond holdings out of our taxable accounts (the common "smart" practice due to the taxable interest they generate), we move more stocks into them--and stocks have higher expected growth rates, so maybe that will prove to be less than optimum (esp if the tax treatment of dividends and cap gains changes). Basically, the hoped-for efficiency we want through this tax fine-tuning is highly sensitive to unknowable future growth rates and unknowable future tax law , and it's likely fairly minor at any rate for people at the 15% bracket.

My gut tells me that if the perceived complexity of properly putting assets into the right category (taxable, tax-deferred, or tax-free) accounts and then rebalancing them annually drives a person to an FA, then they would have been much better off with simple balanced funds in each category (just due to FA fees, saying nothing of other potential pitfalls). And I'm not really sure that those of us who try to put the right kind of asset in each type of account to minimize taxes will ultimately get much for our efforts. But, I'm still doing it . . .

[Jerry1]

Read the article and like the framework. Question - when do you withdraw and when do you rebalance? Let's say you're going to withdraw/spend $100,000. Do you do that on Jan. 1, each year and rebalance at the same time? Or, do you withdraw as you spend throughout the year and rebalance at the end of the year?

To add some tax planning into this, I plan on withdrawing up to the 15% tax bracket from my tax deferred accounts and use my already taxed account for the rest. I think that will work for a few years - maybe 5.

[pb4uski]

I keep ~6% of my nestegg in cash in an online savings account paying 0.95%. When combined with taxable account dividends, this would be sufficient to meet our spending needs for about 3 years. I have a monthly transfer (my "paycheck") from this online savings account to my local bank account from which I pay my bills. I also take dividends in cash and they go directly into my local bank account. I monitor my local bank balances and make additional transfers if needed.

I typically rebalance in December. I sell stocks in my taxable account to bring my cash back up to 6% so I then know my capital gains for the year. Then I do a proforma tax return and calculate a Roth conversion to the top of the 15% tax bracket and make that Roth conversion. Then I complete rebalancing by selling/buying fixed income or equity as necessary within my tax-deferred accounts.

[big-papa]

Quote:

Originally Posted by Sunset Cute view of cash, but really cash is a separate category. If the Feds raised interest rates by 2.5% instead of .25% you would have probably seen all bond funds drop by something. However your cash would have remained exactly the same, but it's potential would have just increased as you could earn more interest in a CD on it.

I agree when you're comparing bond funds to cash. But if you hold actual bonds and hold them to maturity then bonds and cash start to look a lot alike....

[mathjak107]

not really as cash changes quickly in a rising rate scenario . bonds do not .

you will always be behind the curve with bonds when rates rise on both bonds and cash . they don't always move together .

[big-papa]

Quote:

Originally Posted by mathjak107 not really as cash changes quickly in a rising rate scenario . bonds do not . you will always be behind the curve with bonds when rates rise on both bonds and cash . they don't always move together .

You are correct, but still a totally different beast than bond funds. Here's a good article, complete with a 6 month cd - 6 month treasury comparison chart.

https://www.bogleheads.org/wiki/CDs_vs_bonds

[gerntz]

Perhaps I missed it, but I don't see an analysis of taking more retirement distributions before starting SS & RMD's to keep later income/tax rates down, keep Medicare payments down, & to add to Roth IRA's. I.e., it's not good to pay low taxes now if they will be higher later. JMO.

[Trooper]

Quote:

Originally Posted by pb4uski I keep ~6% of my nestegg in cash in an online savings account paying 0.95%. When combined with taxable account dividends, this would be sufficient to meet our spending needs for about 3 years. I have a monthly transfer (my "paycheck") from this online savings account to my local bank account from which I pay my bills. I also take dividends in cash and they go directly into my local bank account. I monitor my local bank balances and make additional transfers if needed. I typically rebalance in December. I sell stocks in my taxable account to bring my cash back up to 6% so I then know my capital gains for the year. Then I do a proforma tax return and calculate a Roth conversion to the top of the 15% tax bracket and make that Roth conversion. Then I complete rebalancing by selling/buying fixed income or equity as necessary within my tax-deferred accounts.

That will be my plan also when I say "Adios" in about 6 weeks, pb4uski.

[pb4uski]

Quote:

Originally Posted by gerntz Perhaps I missed it, but I don't see an analysis of taking more retirement distributions before starting SS & RMD's to keep later income/tax rates down, keep Medicare payments down, & to add to Roth IRA's. I.e., it's not good to pay low taxes now if they will be higher later. JMO.

You missed it. There are numerous threads on the "tax torpedo" and on the strategy of doing annual Roth conversions to the top of a specific tax bracket that is lower than the tax bracket that one expects to be in once SS and RMDs start. Most typically, people do Roth conversions to the top of the 15% tax bracket. Google "tax torpedo" in the search bar above.

[samclem]

Quote:

Originally Posted by pb4uski I also take dividends in cash and they go directly into my local bank account.

Just out of curiosity, why have you chosen to send the divs to your local bank rather than the online checking "accumulation" fund? Obviously, there's no right answer, but I'd probably have elected to send them to the online accumulation account, and have the "paycheck" and other inflows to my spending account be a very regular and fixed amount each month (1/12 th of the withdrawal amount I determined in Dec).

[gerntz]

Quote:

Originally Posted by pb4uski You missed it. There are numerous threads on the "tax torpedo" and on the strategy of doing annual Roth conversions to the top of a specific tax bracket that is lower than the tax bracket that one expects to be in once SS and RMDs start. Most typically, people do Roth conversions to the top of the 15% tax bracket. Google "tax torpedo" in the search bar above.

I meant in this thread, not others. Perhaps the op isn't aware if he/she didn't mention?

[pb4uski]

Quote:

Originally Posted by samclem Just out of curiosity, why have you chosen to send the divs to your local bank rather than the online checking "accumulation" fund? Obviously, there's no right answer, but I'd probably have elected to send them to the online accumulation account, and have the "paycheck" and other inflows to my spending account be a very regular and fixed amount each month (1/12 th of the withdrawal amount I determined in Dec).

No particular reason, that's just the way I did it. It does work out well however in that most of the dividends come in at year end and end up being used for Christmas shopping and estimated tax payments (which I only do in December).

[heyyou]

@ the OP, your plan looks good. I like your willingness to accept the variable income of the flat percentage WD of your recent annual portfolio value. WDing that way is not a problem even for the slightly higher percentage draw during the SS wait. Your method tolerates the modest temporary boost.

It is the 4% of retirement day assets plus inflation (Bengen's 4% real income SWR) method that cannot be boosted permanently early in retirement due to its sequence of return risk. Your acceptance of variable retirement income dodges that risk.

You can somewhat smooth the larger income variations by spending an average of several years of the variable income. With your 2-3 years of income bucket, you could spend the average of those year's WDs, or just calculate the multi-year average if you choose to not have the separate cash bucket.

Wm Bengen did two studies, several years apart. One was the seminal study that made his reputation. The other may have looked at your fixed % of annual assets method for 20-30-40 retirement years but very few liked the idea of having variable real retirement income. You might have to log in at wherever it was published. My suggestion is to research it because you may be able to spend a few fractions of a % more than the 4% real SWR, since you are accommodating the market drops that the 4% real SWR ignores.

[Chuckanut]

Quite a good discussion and a good article by Kitces. Thanks to all.

[samclem]

And, it is probably worth mentioning the "VPW" (Variable Percentage Withdrawal) calculation method in any discussion of this subject. It has many fans. A good thread on it here.

More details are at the Bogleheads site.

[heyyou]

For what it is worth, the VPW site shows the annual portfolio values for a 1966 retiree. As my retirement progresses, it is comforting to compare my history of portfolio values to the worst case of 1966, regardless of which WD method is used.

[Turn_the_Page]

Your plan is very similar to mine, which I'll implement next year.

[JP.mpls]

I want to again thank everyone for their comments.


Samclem:
I planned on rebalancing my allocation, but the Kitces article really emphasized the importance of doing this. Thanks for posting it.
Fritz:
You are giving me confidence that our similar withdrawal strategies have been working out well for you over the last 6 years. Thanks.
LOL:
I suspect using bonds instead of 2 years of cash is a logical approach. I psychologically like the idea of having a cash supply to withdraw from. Your idea makes putting everything into a 60/40 balance index fund at Vanguard and letting it ride very tempting.


I have two more comments/questions:
- I'm going to show my ignorance here. My retirement budget takes into account medical costs, and taxes to cover the money that will be withdrawn from IRA/401K accounts. While reading everyone's comments I started to realize that I'm not taking inflation into account. Am I supposed to consider increasing the withdrawal percentage as we go along? I certainly haven't planned for that, and it has me very concerned.


- I crunched our numbers a few more times, and came to the following conclusion: A 4% withdrawal rate gives us a very generous monthly amount to live on. In reality we can easily live on something closer to 3%, in fact we are living on something close to the 3% rate. If we plan on this lower 3% withdrawal rate, that leaves us with a nice big slush fund for extra expenditures (car replacement, home repairs, vacations). I realize that I'm just playing mind games with the numbers, but I like the lower withdrawal rate plan better. Any comments on this idea?




Thanks. JP